Quality Characteristic Study of Star Fruit (Averrhoa carambola L.)Using an Image Processing Technique

Yusuf Hendrawan, Sumardi H. S.


The research aim was to develop image processing algorithms suitable for the determination of star fruits quality by assessing the external appearances of the fruit i.e. physical characteristics used for grading. Based on that, a computer software for star fruit quality grading was then developed. An RGB (Red-Green-Blue) index was used to determine the colour range and classification, while an euclidian method was used to measure the length and diameter of the fruit to classify the quality grade.

The result showed that an image processing has high correlation between fruit length (r2=0.97), fruit diameter(r2=0.98),and fruit area based on fruit length (r2=0.93), but less correlated with the fruit area based on the fruit diameter (r2=0.62). The red index was slightly well correlated with the total soluble solid (°Brix) of the fruit juice (r2=0.74). On the other hand, the green and blue indexes showed a low correlation with the total soluble solid of the fruit with a correlation value of 0.34 and 0.10 respectively. It can be concluded that the quality characteristics of star fruit may be mostly determined by the red index, fruit length, fruit diameter, and fruit area. The Grade A quality characteristics of the fruit can be expressed with a value of 0.402-0.460 (red index),  186-237 pixels (fruits length), 100-125 pixels (fruits diameter), and  13627-20449 pixels (fruits area). While the values of such parameter of the B Grade were 0.402-0.460, 147-185 pixels, 76-90 pixels, and 8805-13415 pixels, and of C Grade were 0.370-0.401, 147-185 pixels, 76-90 pixels, and 8805-13415 pixels, respectively.


Key Words : Star Fruits, Image Processing, quality.

Full Text:



  • There are currently no refbacks.

Copyright (c) 2019 Yusuf Hendrawan, Sumardi H. S.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.